Not the Powerplay: Overs 7 to 15 Are Asia's Real T20 Baseline

I have an old habit in the Sharjah press box: the moment the first six overs of an innings end, I photograph the scoreboard. Since early 2026 I have collected more than four hundred such photographs. Last month I laid them out side by side and one number stopped me. In 47 matches where a team scored 60-plus in the powerplay, that same team lost. That is roughly 49 percent — a coin toss. Yet in those very matches, the closing line made the powerplay-heavy side favourite in more than 65 percent of cases.

The number was familiar to me. In 2026 at Footballist I built the K League xG baseline because the goals were lying; Jeonbuk Hyundai Motors were scoring 2.11 goals per game against 1.84 xG, and the market treated that surplus as permanent, pricing them too rich away from home. Cricket's runs and wickets sit in exactly the same place. They look clean, but they do not describe the whole process. In Asian T20 cricket the match is written between overs seven and fifteen; the market still puts its money in the first six.

Asia's T20 calendar now runs almost continuously. The 2026 Asia Cup was played in Dubai, Abu Dhabi and Sharjah. The 2026 T20 World Cup rolled out across India and Sri Lanka. In between sits ILT20 in January and February, six teams over five weeks on UAE grounds. Of the 284 Asian T20 matches on my calendar between February and November 2026, 113 were played at venues where neither side holds a genuine home advantage. The crowd belongs to no one, so noise and pitch conditions have to be read separately — otherwise two variables melt into one.

My baseline is simple, but it took time to build. For every ball I log four things: phase (1-6, 7-15, 16-20), bowler type (spin or pace), strike rate, and the batter's position relative to the last wicket. I trust a number only when I can reproduce it on a quiet Tuesday. So I do not touch a coefficient on a sample under twenty matches. In 2026, after the pandemic break, I waited until matchday six before removing the home-advantage coefficient in the K League, because when the stadiums emptied, home advantage stopped hiding behind the crowd. Cricket needs that same patience now.

My tracking table looks like this — Asian venues only, evening matches only, first innings only:

| Phase | Runs/over 2026-24 | Runs/over 2026-26 | Change | |---|---|---|---| | 1-6 | 8.9 | 9.3 | +0.4 | | 7-15 | 7.4 | 7.1 | -0.3 | | 16-20 | 9.8 | 10.4 | +0.6 |

The first thing the table shows: scoring is rising in the last five overs and rising in the powerplay, but falling across the middle nine. The decimals look small, except that phase is roughly 45 percent of all balls in a T20 innings. The market's attention sits at the two ends, where scoring is climbing; the middle, where matches actually turn, is compressing. Process and price are walking in opposite directions — that is the biggest inefficiency of this moment.

Two distinct causes sit behind the middle-overs squeeze, and reading them as one sends the analysis the wrong way. The first is bowling structure: in Asian conditions, spin economy between overs seven and fifteen is now hovering between 6.4 and 6.8, against roughly 7.1 in 2026. Bowlers like Rashid Khan, Wanindu Hasaranga and Noor Ahmad are not only taking wickets, they are pulling strike rates down. The interesting part is that dot balls are rising faster than wickets. In my logs this phase produced 3.1 dots per over in 2026; it produces 3.7 now. Four dots means the innings stalls, and the batter is then forced to take risk in the next over.

The second cause is batting, and it is the more neglected one. Asian line-ups now manufacture powerplay specialists, but the finisher slotted at seven or below has no plan for batting in the middle overs. So the seventh over often traps an innings between a powerplay batter and a finisher. In tournament cricket, with a match every second day, nobody fills that gap in training. To me this is a tactical blind spot, and it is precisely why one type of team keeps profiting in the numbers: a side with a number three or four who can read spin, and who can rotate strike even without boundaries.

One number belongs here, and it is a pillar of my model. In the 2026-26 sample, the correlation between the 7-15 phase run-rate differential and the match result was 0.63. Over the same sample, the powerplay run-rate differential correlated at just 0.31. Powerplay boundary percentage correlated at 0.26. These are coefficients, not causes, and I say so with that caveat attached: a side running two runs per over ahead in overs seven to fifteen wins roughly four times out of six.

I have another objection about the death overs, and it is under-discussed in Asian markets. Judged as a bowling unit, success between overs sixteen and twenty rests mainly on the second seamer — but nobody prices the economy of the third and fourth seamer. On an evening Dubai pitch, once the ball loses its hardness, the absence of a third seamer who can land the yorker returns five or six runs straight back. This is not a story about individual skill, it is squad construction. In my model I keep 'third seamer economy, death overs' as a separate variable, because in a knockout that column is worth twenty runs.

An environmental adjustment box belongs here, because I put one in every preview. At Dubai International Stadium in the evening, the average run-rate gap between the first six overs and overs seven to fifteen across a sample of 875 matches is about 2.1. In Sharjah the gap is 1.6, because the ground is small and the boundary rope sits further back. In Abu Dhabi, under floodlights, I have watched spin revolutions drop across the last ten overs of the first innings, and capturing that in the model cuts prediction error on economy by 0.4. These adjustments are not reader-friendly, but the market does not reconcile them either.

Workload and travel add another layer, and in Asia's tournament reality that layer is thick. In ILT20, even when matches are at the same venue on consecutive nights, the rest between innings is minimal. By my count, a bowling unit playing back-to-back nights concedes roughly 0.4 more per over in the 7-15 phase in the following match, rising to 0.7 in the death overs. Nobody changes strategy over that amount, but if two or three of your matches fall in that gap during a tournament, it is two points on the table. Travel and schedule density are context to me, not emotion.

Now I throw the counter-question at myself, because my loyalty to the model is not blind. If the powerplay matters so little, what happens if every side abandons the first six overs and pours resources into the middle? The answer is not simple, and this is where the wall between correlation and causation stays standing. A big powerplay often arrives against a weak bowling attack, so the number reflects the opponent's weakness, not the batting side's strength. The reverse can also hold: a team that loses no wickets in the powerplay can take fewer risks through the middle, and its economy looks better there. The causal chain does not run in one direction only. Kazan reminded me that a model can be right and still lose; in 2026 the market priced Germany -1.5 at 78 percent implied probability while the performance data said otherwise. The model was correct, the result arrived differently. So I now pre-register outcome ranges and review calibration afterwards.

There is one more trap I have walked into myself: taking large positions on the back of these decimals in thin markets. Across several Asian bilateral series and domestic leagues the closing line is itself suspect, and chasing 'value' there means matching your price to a wrong one. My rule has three parts — sufficient liquidity, closing-line value, and a minimum sample. If all three do not hold, the analysis stays written and the position stays unopened. In thin markets, edge is not a signal; it is noise.

So what do you watch next? In the next tournament, take the first six matches and log the 7-15 phase economy separately. Then look at where that number lands once the full season is done. My expectation: if strong batting line-ups start sending a pace-breaker in at the seventh over, the middle-phase run rate will climb again — and the market will reprice, late. But I will not change a coefficient after one weekend. Let twenty matches finish first. Then we talk.